🚧 Fill empty l1 records with l2 aggregates

Database/fill_data_gap.py

@@ -159,3 +159,21 @@
                             )
             except BaseException as err:
                 logger.error(f"Could not fill area data gap for {impact} at l3. Error: {err}")
+
+    logger.info("Filling data gap upwards (l2 -> l1) for NULL impacts")
+    for impact in l2.keys():
+        empty_l1_events = l1[
+            [dg_util.event_id, f"Total_{impact}_Min", f"Total_{impact}_Max", f"Total_{impact}_Approx"]
+        ][l1[f"Total_{impact}_Min"].isna()][dg_util.event_id].unique()
+        for e_id in empty_l1_events:
+            impact_per_event_id = l2[impact][[dg_util.num_min, dg_util.num_max]][
+                (l2[impact][dg_util.event_id] == e_id) & (~l2[impact][dg_util.num_min].isna())
+            ]
+            if not impact_per_event_id.empty:
+                agg_min, agg_max = impact_per_event_id.sum()
+                l1[l1[dg_util.event_id] == e_id] = l1[l1[dg_util.event_id] == e_id].apply(
+                    lambda row: dg_util.l2_to_l1(row, agg_min, agg_max, impact), axis=1
+                )
+                logger.info(
+                    f"Aggregated values found for {e_id} in impact category {impact}: agg_min = {agg_min}, agg_max = {agg_max}"
+                )

Database/scr/normalize_data.py

@@ -79,6 +79,13 @@ def fill_area(self, row: dict, missing_areas: dict[str, list], area_col: str) ->
             row[f"{self.admin_areas}_{c}"] = row[f"{self.admin_areas}_{c}"].extend(missing_areas[f"{area_col}_{c}"])
         return row
 
+    @staticmethod
+    def l2_to_l1(row: dict, agg_min: float, agg_max: float, impact: str) -> dict:
+        row[f"Total_{impact}_Min"] = agg_min
+        row[f"Total_{impact}_Max"] = agg_max
+        row[f"Total_{impact}_Approx"] = 1
+        return row
+
     def l3_to_l2(self, l3_row: dict) -> dict:
         l2_row = {}
         for k in l3_row.keys():